On 8/15/07, Dr.Ruud <[EMAIL PROTECTED]> wrote:
> "Chas Owens" schreef:
>
> > A one second sleep is common, but you can sleep for less than a second
> > with usleep from the Time::HiRes* module if this is too slow for your
> > purposes.
>
> AFAIK, a "sleep 1" could take anything between 0 and 2 seconds. See also
> select().
snip
Well, technically, a "sleep 1" could take over an hour of wall time;
however, the request is for one second (some old machines used the
next second which was often less than one full second) and it is
usually one second (or so near as to not matter). With no load on my
system I get skew of less than .01 seconds per iteration of
perl -MTime::HiRes -le '
while (1) {
print join ".", Time::HiRes::gettimeofday;
sleep 1;
}
'
and I believe most of that is the is the code running between the
calls to sleep. I get the same skew with a pure cpu load (2 busy
waits for my 2 cpu system). Even the number (I used 3000) of greedy
processes doesn't change the skew even though I lose my mouse pointer
in X.
from perldoc -f sleep
On some older systems, it may sleep up to a full second less
than what you requested, depending on how it counts seconds.
Most modern systems always sleep the full amount. They may
appear to sleep longer than that, however, because your process
might not be scheduled right away in a busy multitasking sys‐
tem.
For delays of finer granularity than one second, you may use
Perl's "syscall" interface to access setitimer(2) if your sys‐
tem supports it, or else see "select" above. The Time::HiRes
module (from CPAN, and starting from Perl 5.8 part of the stan‐
dard distribution) may also help.
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
http://learn.perl.org/
